The language of ancient Egypt represents an autonomous branch of one of the most widespread language families in the world, variously called Afroasiatic, Hamito-Semitic, or Semito-Hamitic and comprising, from antiquity to the present time, the entire area of the eastern Mediterranean, northern Africa, and western Asia. The six individual branches of the Afroasiatic family are Ancient Egyptian, Semitic, Berber, Cushitic, Chadic, Omotic. Ancient Egyptian shows the closest relations to Semitic and Berber, more distant ones to Cushitic (except Beja, a Northern Cushitic language possibly derived from the language of the people called mḏ3w in Egyptian texts) and Chadic. Afroasiatic languages are generally characterized by the following linguistic features not all of which are shared by Egyptian: the presence of bi- or triconsonantal lexical roots; a consonantal system displaying a series of ejective, emphatic, or glottalized phonemes (sound units) alongside the voiced and the voiceless series; a vocalic system originally limited to the three phonemes /a/ /i/ /u/; a nominal feminine suffix ⋆-at; a rather rudimentary and often lexicalized case system; a nominal prefix m-; an adjectival suffix -ī, called nisba; a basic opposition between prefix (dynamic) and suffix conjugation (stative) in the verbal system; a conjugation pattern 1st sg. ⋆'a-, 2d ⋆t-, 3d masc. ⋆y-, fem. ⋆t-, 1st pl. ⋆n-, with supplementary suffixes in the 2d and 3d fem. pl.
The history of the Egyptian language, which remained in use over more than four millennia (c. 3000 BCE–1300 CE), can be divided into two main stages, characterized by a major typological change from synthetic to analytic patterns in the nominal syntax and in the verbal system. Each of these two stages of the language can be further subdivided into three different phases, affecting primarily the sphere of graphemics.
The forms of Older Egyptian consisted of the language of all written texts from 3000 to 1300 BCE, surviving in formal religious texts until the second century CE. It has three main phases: Old Egyptian, Middle Egyptian, and Late Middle Egyptian.
The language of the Old Kingdom and the First Intermediate Period (3000–2000) is known as Old Egyptian. The main documents of this stage of the language are represented by the religious corpus of the Pyramid Texts and by a sizable number of so-called autobiographies on the external walls of the rock tombs of the administrative elite.
Also termed Classical Egyptian, Middle Egyptian lasted from the Middle Kingdom to the end of the eighteenth dynasty (2000–1300). Middle Egyptian is the classical language of Egyptian literature, expressed in a variety of texts that can be classified according to two main genres. The first, the “Instructions,” are wisdom texts normally addressed from a father to a son, which conveyed the educational and professional expectations of Egyptian society. The most renowned examples are the Instructions of the Vizier Ptahhotep and the Instructions for Merikare. Some of these moral texts, such as the Admonitions of Ipu-Wer, are in fact philosophical discussions ex eventu on the state of the country taking as a point of departure the political evolution from the Old to the Middle Kingdom, a period generally refered to as the First Intermediate Period. The second, the Tales, are narratives relating the adventures of a specific hero and representing the vehicle of individual, as opposed to societal, concerns. The most famous specimens of this genre are the Tale of Sinuhe and the Shipwrecked Sailor. Some texts, such as the Eloquent Peasant, combine features and contents of both main genres.
Late Middle Egyptian.
In use from the late New Kingdom to the end of Egyptian civilization, Late Middle Egyptian is the language of mainly religious texts (rituals, mythological inscriptions, hymns). It maintains the linguistic structures of the classical language, but especially in the Greco-Roman period (Ptolemaic Egyptian: third century BCE–second century CE), it shows an enormous extension in its set of hieroglyphic signs.
Older Egyptian is characterized by its preference for synthetic grammatical structures. It displays a full set of morphological suffixes indicating gender and number; it exhibits no definite article: rmṯ “the man,” “a man”; it maintains the verb-subject-object order in verbal formations (sḏm=k n=f “may you listen to him”).
The forms of Later Egyptian are documented from the nineteenth dynasty to the Middle Ages (1300 BCE–1300 CE) and consisted of Late Egyptian, Coptic, and Demotic.
In use from 1300 to 700 BCE, Late Egyptian was the language of written records from the second part of the New Kingdom. It primarily conveys the rich entertainment literature of the nineteenth dynasty, consisting of both traditional forms, for example, the Tale of the Two Brothers, the Tale of Wenamun, or the Instructions of Ani and the Instructions of Amenemope, and of new literary genres, such as the mythological tale or love poetry. Late Egyptian was also the idiom of the Ramesside bureaucracy: for example, archival documents from the Theban necropolis or school texts such as the Miscellanies.
In use from the seventh century BCE to the fifth century CE, Demotic was the language of administration and literature during the Late Period (seventh–fourth century BCE). Although grammatically it closely continues Late Egyptian, Demotic differs radically from it in its graphic system. Important texts in Demotic are the narrative cycles of Setne-Khaemwase and of Petubastis and the instructions of Papyrus Insinger and Onkhsheshonqi.
In use from the fourth through fourteenth centuries CE, Coptic was the language of Christian Egypt, written in a variety of the Greek alphabet completed by six or seven Demotic signs to indicate Egyptian phonemes absent in Greek. As a spoken language Coptic was superseded by Arabic from the ninth century, but it survives to the present in the liturgy of the Coptic Orthodox Church of Egypt.
As opposed to Older Egyptian, Later Egyptian developed analytic features. Suffixal markers of morphological oppositions are dropped and functionally replaced by prefixal morphemes, the demonstrative “this” and the numeral “one” evolving into the definite and the indefinite article respectively. Periphrastic patterns in the order subject- verb-object supersede older verb-subject-object verbal formations.
Unless otherwise specified, the following grammatical sketch refers to Older Egyptian, more specifically to the language of classical Middle Egyptian literature.
The language of Ancient Egypt was written in a monumental script and its cursive varieties. The monumental script is known as hieroglyphs. These are pictographic signs that represent living beings and objects, such as gods or people, animals, parts of the human or animal body, plants, stars, buildings, furniture, containers, and so forth. The number of hieroglyphic signs varied from approximately one thousand in the Old Kingdom down to about 750 in the classical language, but—following the decline of a centralized school system—increased dramatically to many thousands during the Ptolemaic and Roman rule in the country.
In this writing system, phonological and ideographic principles appear combined: a written word consists of a sequence of signs, called phonograms, which convey a substantial portion of its phonological structure: normally all the consonants, occasionally also the semivocalic phonemes. Each phonogram can express one, two, or three consonants of the language. The sequence of phonograms is usually followed by a “determinative,” which indicates iconically the semantic sphere of the word: for example, a sitting god expresses the lexical sphere of “god, divine,” a scribe's outfit indicates the semantic realm of “writing,” a stylized desert landscape denotes the word as a foreign toponym. Although some words in common use may be written only with phonograms, many items of the basic vocabulary are expressed by hieroglyphic signs that represent, evoke, or symbolize their own meaning: these are called logograms or ideograms: for example, the hieroglyphic sign representing a human head is used to signify the lexeme “head.” The use of a hieroglyphic sign as logogram or ideogram is made visually explicit through a vertical stroke following the sign.
As part of its inventory of phonograms, hieroglyphic writing displayed a set of twenty-four “alphabetic” signs that expressed one consonantal phoneme. Each of these signs corresponded to one consonantal or semiconsonantal phoneme of Egyptian; in this way, we can obtain an insight into the complete phonological system of the language, the only exception being the consonant /l/, for which a distinct grapheme appears only in Demotic. In spite of the presence of this exact correspondence between monoconsonantal signs and the phoneme inventory, hieroglyphs never developed a genuine alphabetic system. This was due to the culturally significant connection between the hieroglyphic sign and its pictographic content, a connection that in the later periods of the history of the hieroglyphic system, especially in Ptolemaic Egyptian, led to the emergence of hitherto unknown signs, of new phonological values for known signs, and of cryptographic solutions.
The hieroglyphic system was mostly used for monumental purposes, and only rarely in a cursive form adopted in religious texts of the Middle and the New Kingdom. During the three millennia of their productive use, hieroglyphs developed two manual forms: while Hieratic (2600 BCE–third century CE) represents a direct cursive rendering, with ligatures and diacritic signs, of individual hieroglyphs, Demotic (seventh century BCE–fifth century CE) modifies completely the writing conventions by introducing a simplification of hieratic sign-groups. In Coptic, the hieroglyphic-based system is replaced by an alphabet derived from the Greek one, with the addition of six or seven Demotic signs for the indication of phonemes absent from Greek.
The basic orientation of the Egyptian writing system, and the only one used in the cursive varieties, is from right to left; in epigraphic, that is, monumental texts, this order is often inverted for reasons of symmetry or artistic composition.
Table 1. Consonants of Older Egyptian
|Voiceless||p/p/||t/t/||ṯ/c/||k /k/||ḳ, q /q/||ỉ/ʔ/|
|Voiced||b /b/||d /d/||ḏ /ɬ/||g /g/|
|Voiceless||f /f/||s /s/||š /š/||ẖ /ç/||ẖ /x/||ḥ /h/||h /h/|
|Voiced||z /z/||῾ /ʔ/|
|Nasals||m /m/||n /n/|
|Lateral||3, n, r /l/|
|Vibrant||r /R/||3 /R/|
|Glides||w /w/||y, jj /j/|
The exact phonological value of many Egyptian consonants is obscured by difficulties in establishing reliable Afroasiatic correspondences: Vocalism and prosody can only be reconstructed by combining the contemporary Akkadian transcriptions of the second millennium with the later Greek transcriptions of the Late Period (corresponding roughly to spoken Demotic) and the Coptic evidence of the first millennium CE.
For Older Egyptian, we can posit the phonological inventory shown in table 1 (the conventional Egyptological transcriptions are indicated in italics). The Egyptian phonological system does not display the “emphatic” phonemes common to many Afroasiatic languages. Voiced stops, however, were often articulated as “glottalized” consonants; thus, /d/ = [t'], /j/ = [c'], /g/ = [k']. The phoneme conventionally transcribed 3 (aleph), originally a uvular vibrant probably articulated like r (grasseyé) in French, corresponds to Afroasiatic ⋆r or ⋆l. The existence of a phoneme /l/ in Egyptian is well established on the basis of Afroasiatic correspondences and of its presence in Coptic; yet, unlike all other consonantal phonemes, /l/ shows no unequivocal graphic rendering, being expressed in different lexemes by 〈3〉, 〈n〉, 〈r〉 or 〈nr〉.
The opposition between voiced and voiceless consonants tends to be progressively neutralized into the voiceless variant. During the second millennium, the following sound changes took place: (a) the uvulat vibrant /R/ acquired the articulation /ʔ/ (glottal stop); (b) the point of articulation was progressively moved to the front—velar to palatal, palatal to apical; (c) oppositions between fricatives in the palatal region (/š/, /ç/, /x/) tended to be neutralized into /š/; and (d) final /r/ and /t/ tended to become /'/, then to disappear. During the first millennium, the opposition between ῾ayin /ʕ/ and ῾aleph /ʔ/ was also neutralized.
The original set of vowels (see table 2) underwent a certain number of historical changes. Already during the second millennium, short stressed /i/ and /u/ merged into /e/, and long /u:/ turned into /e:/. Around 1000, long /a:/ became /o:/, a phonetic evolution similar to the contemporaneous “Canaanite shift” in Northwest Semitic. There was also a change in the short tonic from /a/ to /o/, but Coptic shows that this affected only a portion of the Egyptian linguistic domain. Egyptian unstressed vowels progressively lost phonological status and became realized as schwa (ǝ).
In the classical language, the stress could lie only on the ultimate or penultimate syllable. Although both closed and open syllables are found in pretonic position, the only possible structure in syllables following the stress is the closed syllable with short vowel. The stressed vowel of a penultimate open syllable is long; some scholars posit the existence of extralong syllables under oxytone stress. In addition to final /j/, /w/, /t/ and /r/, unstressed vowels were dropped between 2000 and 1600.
The basic Egyptian morphological unit of nouns is a biliteral or triliteral root, which is modified by flectional suffixes (see table 3). Although some scholars detect remnants of the Afroasiatic case system in the Egyptian verbal paradigm, the syntactic structure of classical Egyptian was so rigid that cases could not have played a productive role in the language in historical times. Short unstressed vowels in final position are not posited for classical Egyptian.
Adjectives are morphologically and syntactically treated like substantives. Very common is the derivational pattern called nisbation, in which a morpheme -j is added to a noun to form its corresponding adjective: nṯr, “god,” nṯrj “divine.”
The three types of pronouns are listed in table 4. Independent pronouns are used for the topicalized subject of sentences with nominal or adjectival predicate in the first person (jnk jtj=k “I am your father”) and for the focalized subject of nominal or verbal cleft sentences: ntf z3 Wsjr “he is Osiris' son”; jnk jnj=j sw “it is I who shall bring it.” Dependent pronouns are used as the object of a transitive verb (sḏm=f wj “he will hear me”), as the subject of a qualifying nominal sentence (nfr ṯw ḥn῾ = j “you are happy with me”) and of an adverbial sentence in the first and second persons only after an initial particle (mk wj m-b3ḥ=k “Look, I am in front of you”). Suffix pronouns are used as the subject of verbal forms, as possessive pronoun, and as object of prepositions: ḏj = k r = k n = j ḫ.t = j “you shall truly (lit., to-you) give me (= to-me) my possessions.”
Table 2. Vowels of Older Egyptian
Table 3. Nominal Morphology in Older Egyptian
|Plural||.ϕ, .w, .ww||.t, .jjt, .wt|
Demonstratives are characterized by a deictic (specifying) element preceded by the indicator of gender and number (masc., pn, pf, pw; fem., tn, tf, tw; rmṯ pf “that man”; ḥjm.t tn “this woman”). They normally follow the noun to which they refer. The plurals nn, nf, nn are also used with following genitive as pronouns in partitive constructions: nn nj srjw.w “these officials.” Later Egyptian developed a set of definite articles from the proclitic demonstrative series p3, t3, pl. n3; p3-rm(t) “the man”; t3-ḥm(.t) “the woman”; n3-srj.w “the officials,” as well as an indefinite article from the numeral w῾j “one”: w῾-rmt “a man.”
The relative pronoun is masc. ntj; fem. nt.t; pl. ntj.w “who/which/that.” It only refers to specific (i.e., semantically determined) antecedents because in Egyptian indefinite antecedents are not resumed by a relative, but by a circumstantial, clause: rmṯ ntj rḫ.n=j sw “the man whom I know” (lit., “man that I know him,” as opposed to rmṯ rḫ.n=j sw “a man that I know,” lit., “man—I know him”). Peculiar to Egyptian is the presence of a relative pronoun, which semantically incorporates negation: masc. jwtj, fem. jwt.t, pl. jwtj.w “who/which/that not.”
Table 4. Personal Pronouns in Older Egyptian
|2 m.||ntk, ṯwt1||yṯ w||=k|
|2 f.||ntṯ, ṯmt1||ṯn||=ṯ|
|3 m.||ntf, swt1||sw||=f|
|3 f.||nts, stt1||sj, st||=s|
|3 c.||ntsn||sn, st||=sn|
|1 twt, ṯmt, swt, and stt are archaic forms found mainly in Old Kingdom religious texts.|
Basic interrogative pronouns are m “who?”, “what?”, jḫ “what?”, jšst “what?”. They can be combined with prepositions or particles to form complex pronouns: jn-m “who?”, literally, “FOCUS-who”; ḥr-m “why?”, literally, “on- what?”.
The most common numerals show etymological connections with other languages of the Afroasiatic family: w῾j “one,” sn.wj “two,” ḫmt.w “three,” jfd.w “four,” dj.w “five,” srs.w “six,” sjḫ.w “seven,” ḫmn.w “eight,” psḏ.w “nine,” mḏw “ten,” ḏw.tj “twenty,” m῾b3 “thirty,” ḥm “forty,” ty.w “fifty,” sjsy.w “sixty,” sfḫy.w “seventy,” ḫmny.w “eighty,” psḏy.w “ninety,” šn.t “one hundred,” ḫ3 “one thousand,” ḏb῾ “ten thousand,” ḥfn “one hundred thousand,” ḥḥ “one million.” It is noteworthy that five is etymologically a nisba of Afroasiatic ⋆yad “hand,” which is no longer a productive word in historical Egyptian, that “twenty” is an old dual of “ten,” that “thirty” and “forty” are independent lexemes but cardinals from “fifty” to “ninety” represent the plural forms of the respective units “five” through “nine.” Numbers are seldom written out; they are mostly rendered graphically, strokes indicating the units, and special hieroglyphs being used for the powers of “ten.” Ordinals are derived from cardinals through the addition of a suffx. nw (from 2 to 9, e.g., ḫmt.nw “third”), or the prefixation of the participle mḥ “filling” to the cardinal number (from 10 onward, e.g., mḥ-20 “twentieth”).
Finite verbal forms are built by annexing a suffix pronoun to the root, either directly or after a morpheme indicating tense, aspect, or voice features. The most important verbal indicators are .n (past tense; sḏm.n=j “I heard”), .t (perfective, sometimes prospective aspect, n sḏm.t=f “he had not heard,” jw.t=f “he shall come”), .w (prospective aspect and passive voice, jrj.w=f “it has been/shall be done), .tw (passive voice, sḏm.tw=k “you are heard”). Classes of so-called “weak” verbal roots, whose third radical is a semiconsonantal j or w, show the reduplication of the second radical and the presence of a stressed vowel between the two consonants in a form indicating in Semitic languages the imperfective aspect or the factitive stem, such as the imperfective in Akkadian and the D-stem throughout Semitic: In Egyptian this form, which is conventionally called emphatic, fulfills the function of pragmatic “theme” of the sentence in which it appears: mrr=s wj “(the-fact-that-)she loves me.” In such sentences, the attentional stress normally lies on an adverbial modifier. The nonreduplicated verbal form is further subdivided into temporally unmarked or future functions, depending on the temporal setting of the context. The imperative has no suffix element in the singular, but sometimes, especially with weak verbs, a suffix .w or .y in the plural.
Egyptian also exhibits a verbal form, variously called old perfective, stative, or pseudoparticiple, which indicates the wide semantic range of verbal “perfectivity,” from perfective aspect with intransitive verbs to passive voice with transitive verbs. This form displays a set of suffix pronouns that are etymologically linked to the forms of the Semitic suffix conjugation. Examples are prj.kw “I have come forth,” ḫpr.tj “you have become,” and rḏj.w “it has been given.”
Nonfinite forms of the Egyptian verb are (1) the participle (nomen agentis), which exhibits nominal morphology and is derived from the verbal stem (2) the infinitive (nomen actionis), which shows a suffix .ø in the regular verbs and a suffix .t in some classes of weak verbs (mrj.t “to love,” “love”). A special type of infinitive characterized by a suffix .w is used after verbs of negative predication, such as tm “not to do.” It is a matter of scholarly dispute whether the radical element in finite forms was originally a nomen agentis (mrj=f “⋆a-loving-one-is-he” > “he loves”) or a nomen actionis (mrj=f “loving-of-him” > “he loves”). Verbal predications could also be expressed by prepositional constructions (sw ḥr sḏm “he is on hearing” > “he is hearing”; jw=f r mrj.t “he is toward loving” > “he is going to love,” “he will love”); constructions such as these characterize the evolution of the verbal system in Later Egyptian.
The most frequent prepositions are m “in,” “with”; n “to,” “for”; r “toward”; mj “as,” “like”; ḥr “on”; ẖr “under”; ḥn῾ “with”; ḫft “according to”; and ḫntj “before.” Prepositional phrases follow the noun or the verb they modify. Particularly noteworthy is the presence of the preposition ḫr “near”, whose original semantic value (A ḫr B “A is near B”) was applied to any situation in which the two participants A and B find themselves at different hierarchical levels, A being socioculturally higher than B (sḏd=f ḫr msj.w “he will speak to the children”) or vice versa B higher than A (jm3ḫy ḫr nṯr ῾3 “honored by the great god”).
The basic negative particle is n (cf. Semitic ⋆la'): n rḫ.n=f “he does not know”: A variant of this, conventionally transcribed nn, is used as predicative negation: nn m3῾.tjw “there are no trustworthy people.”
Egyptian exhibits three sentence types, nominal, adverbial, and verbal, which are classified on the basis of the syntactic nature of their predicate—the subject being always a nominal phrase.
- 1. Nominal sentences, in which the predicate is either a substantive or an adjective. In categorical statements or qualifying adjectival sentences, the normal order of constituents is predicate-subject: R῾w pw “this is Ra”; nfr mtn=j “my path is good.” This sequence is modified into subject-predicate in identifying sentences when both the subject and the predicate are semantically determined or specific (sn.t=f spd.t “his sister is Sirius”), or in cleft sentences, when the subject is focalized (jn sn.t=f s῾nḫ rn=f “it is his sister who causes his name to live”). Topicalization is achieved by means of extraposition, the topicalized syntagm being resumed by a coreferential pronoun in the main clause.
- 2. Adverbial sentences, in which the predicate is an adverbial or prepositional phrase; the order is always subject-predicate: jw NN jr p.t “NN is towards heaven”; ẖr.t=k m prw=k “your rations are in your house.”
- 3. Verbal sentences, with verbal phrase; in which the order is predicate-subject (-object): ḫ῾.y=k “you shall appear”; ḫ῾m.n=f wj “he charged me.”
In Egyptian sentences, verbal forms often appear embedded as theme or rheme of the utterance. In the Egyptological literature, this phenomenon is known as “transposition.” The theme-rheme sequence is labeled “complex adverbial sentence”: wgg 3s.n=f wj “(as for) weakness, it has seized me” > “weakness has seized me”. Thus, Egyptian syntax displays a comparatively high degree of topicalization and focalization phenomena. The bare verbal sentence with the sequence predicate-subject appears less frequently than in related languages, being limited to modal (including prospective) functions.
[See also Hieroglyphs.]
- Allen, James P. The Inflection of the Verb in the Pyramid Texts. Bibliotheca Aegyptia, vol. 2. Malibu, 1984.
- Černý, Jaroslav, and Sarah I. Groll. A Late Egyptian Grammar. 3d ed. Studia Pohl, Series Maior, vol. 4. Rome, 1984.
- Doret, Éric. The Narrative Verbal System of Old and Middle Egyptian. Cahiers d'Orientalisme, vol. 12. Geneva, 1986.
- Edel, Elmar. Altägyptische Grammatik. 2 vols. Analecta Orientalia, vols. 34–35. Rome, 1955–1964.
- Englund, Gertie, and Paul J. Frandsen, eds. Crossroad: Chaos or the Beginning of a New Paradigm; Papers from the Conference on Egyptian Grammar, Helsingør, 28–30 May 1986. Carsten Niebuhr Institut Publications, 1. Copenhagen, 1986.
- Erman, Adolf, and Hermann Grapow. Wörterbuch der ägyptischen Sprache. 7 vols. Leipzig and Berlin, 1926–1963.
- Faulkner, Raymond O. A Concise Dictionary of Middle Egyptian. Oxford, 1962.
- Frandsen, Paul J. An Outline of the Late Egyptian Verbal System. Copenhagen, 1974.
- Gardiner, Alan H. Egyptian Grammar: Being an Introduction to the Study of Hieroglyphs. 3d ed. Oxford, 1957.
- Johnson, Janet H. The Demotic Verbal System. University of Chicago, Oriental Institute, Studies in Ancient Oriental Civilization, no. 38. Chicago, 1976.
- Junge, Friedrich. Syntax der mittelägyptischen Literatursprache. Mainz, 1978.
- Loprieno, Antonio. Das Verbalsystem im Ägyptischen undim Semitischen: Zur Grundlegung einer Aspekttheorie. Göttinger Orientforschungen, vol. IV.17. Wiesbaden, 1986.
- Loprieno, Antonio, ed. Crossroads II: Proceedings of the Second International Conference on Egyptian Grammar, Los Angeles, October 17–20, 1990. Lingua Aegyptia, vol. 1. Göttingen, 1991.
- Ossing, Jürgen. Die Nominalbildung des Ägyptischen. 2 vols. Mainz, 1976.
- Polotsky, Hans Jakob. Collected Papers. Jerusalem, 1971.
- Polotsky, Hans Jakob. “Les transpositions du verbe en égyptien classique.” Israel Oriental Studies 6 (1976): 1–50.
- Polotsky, Hans Jakob. Grandlagen des koptischen Satzbaus. 2 vols. American Studies in Papyrology, vols. 28–29. Decatur, Ill., 1988–1990.
- Schenkel, Wolfgang. Einführung in die altägyptische Sprachwissenschaft. Darmstadt, 1990.
- Shisha-Halevy, Ariel. Coptic Grammatical Categories. Analecta Orientalia, vol. 53. Rome, 1986.
- Vernus, Pascal. Future at Issue: Tense, Mood, and Aspect in Middle Egyptian; Studies in Syntax and Semantics. Yale Egyptological Studies, vol. 4. New Haven, 1990.